The Effect of Labeling Strategies
on the Consumer’s Purchase Intention
The impact of promoting a product using crowdsourcing labeling versus
consumer reviews, top sales or public figure endorsement
Ana Rita Roque
Dissertation written under the supervision of Professor. Claudia Costa, PhD
Dissertation submitted in partial fulfilment of requirements for the
Page: 1 of 64
ABSTRACT
As firms fight to mitigate the effects of uncertainty involved in the process of creating a new product, crowdsourcing offers a promising route as it constitutes one of the most efficient decision-making mechanisms, helping marketers to set apart their products from competitors. Innovation scholars identified the benefits of labelling products as crowdsourcing (user-designed) at the point of purchase. Evidence shows that consumers prefer products labeled as user-designed as opposed to company design, as they are associated with more innovation. Yet, little is known about consumer behavioural intentions when other labeling strategies, such as customer reviews, top sales and public figure endorsement, are presented to determine if any of these have a bigger impact, than crowdsourcing, in the consumer purchase intention. Our findings suggest that none of these labeling strategies is more efficient at the point of purchase, showing that the way the product is communicated to the broader market does not influence consumer´s behavioural intentions. Second, the results showed that the level of perceived user involvement associated to each strategy is also not enough to affect consumer´s behavior.
Enquanto as empresas lutam para mitigar os efeitos da incerteza envolvidos no processo de criação de um novo produto, o crowdsourcing tem sido reconhecido como um caminho promissor, na medida que constitui um dos mecanismos de decisão mais eficientes, auxiliando os especialistas em marketing a distinguirem os seus produtos dos da concorrência. Os especialistas em inovação já começaram a estudar o efeito de rotular produtos como resultado de crowdsourcing (criado pelo consumidor) no ponto de venda. Foi já provado que os consumidores preferem produtos rotulados como criados pelo consumidor, por oposição aos rotulados como criados pelas empresas, na medida em que lhes atribuem maior inovação. Contudo, pouco se sabe ainda sobre as intenções comportamentais do consumidor quando outras estratégias de rotulagem, tais como comentários de consumidores, top de vendas ou recomendação por figura pública, são comparadas com o crowdsourcing, para determinar se alguma terá um maior impacto na intenção de compra do consumidor. Os resultados que alcançámos sugerem que nenhuma destas estratégias de rotulagem é mais eficaz que as outras no ponto de venda, demonstrando que a forma como o produto é comunicado ao mercado não influencia as intenções comportamentais do consumidor. Em segundo lugar os resultados demonstraram que o nível percepcionado de envolvimento do utilizador associado a cada estratégia também não é suficiente para afectar o comportamento do consumidor.
Keywords: Labelling strategies, crowdsourcing, costumer review, top sales, public figure
Page: 2 of 64
TABLE OF CONTENTS
1. INTRODUCTION ... 5
2. LITERATURE REVIEW ... 9
2.1 DEFINITIONS ... 9 2.1.1 Crowdsourcing ... 9 2.1.2 Consumer Reviews ... 11 2.1.3 Top Sales ... 132.1.4 Public Figure Endorsement ... 14
3. HYPOTHESES ... 16
3.1 CROWDSOURCING HYPOTHESIS ... 16
3.2 CONSUMER REVIEW HYPOTHESIS ... 17
3.3 TOP SALES HYPOTHESIS ... 18
3.4 PUBLIC FIGURE ENDORSEMENT HYPOTHESIS ... 19
4. METHODOLOGY ... 20
4.1 OBJECTIVES AND OVERVIEW ... 20
4.2 METHOD ... 20
4.3 SURVEY DESIGN... 20
4.4 THE MEASURES ... 22
5. RESULTS ... 25
5.1 DESCRIPTION OF THE SAMPLE... 25
5.2 PRODUCT INVOLVEMENT ... 25
5.3 MAIN ANALYSIS: TESTING THE HYPOTHESES ... 25
5.3.1 Results for the One Way ANOVAS ... 25
5.3.2 Results for the Regression Analysis ... 28
5.4 FURTHER ANALYSIS ... 29
5.4.1 Product Users Involvement ... 29
5.4.2 New Group: high and low user’s involvement assumption ... 32
6. DISCUSSION AND CONCLUSION... 34
6.1 ACADEMIC IMPLICATIONS ... 36
6.2 MANAGERIAL IMPLICATIONS ... 38
Page: 3 of 64
SOURCES ... 40
APPENDIX ... 47
LIST OF FIGURES
Figure 1 The Conceptual Model ...9Figure 1.1 The Conceptual Model II ...36
LIST OF TABLES
Table 1 Measures ...22Table 2 Measures for Purchase Intention ...23
Table 3 Scenarios on the Purchase Intention ...26
Table 4 Means Scenarios on the Perceived Innovation ...27
Table 4.1 Scenarios on the Perceived Innovation ...27
Table 5 Means Scenarios on the Perceived Quality ...27
Table 5.1 Scenarios on the Perceived Quality ...27
Table 6 Coefficients: Scenarios on the Product Purchase Intention ...28
Table 7 Coefficients: Scenarios on the Attitude Towards the Firm ...28
Table 8 Coefficients: Scenarios on the Perceived Innovation ...29
Table 9 Coefficients: Scenarios on the Perceived Quality ...29
Table 10 Scenarios on Users Involvement ...30
Table 10.1 Scenarios on Users Involvement ...30
Table 11 Users Involvement on Perceived Innovation ...31
Table 11.1 Users Involvement on Perceived Quality ...31
Table 12 Coefficients: Users Involvement on Perceived Innovation ...31
Table 13 Coefficients: Users Involvement on Perceived Quality ...32
Table 14 High and Low Users Involvement on the Purchase Intention ...33
Table 15 High and Low Users Involvement on the Perceived Innovation ...33
Page: 4 of 64
ACKNOWLEDGEMENTS
First of all, I would like to thank my supervisor Professor Claudia Costa, for her valuable guidance, support, and feedback, throughout the dissertation semester, making it possible to develop this thesis.
Also, I would like to express my special and profound thankfulness to my beloved parents for their support, patience, time and incentive to never give up and for showing me that there is always a positive perspective even when results played tricks on me.
I thank Professor Sergio Moreira for his availability and quick answers in the clarification of some questions related to the use of Qualtrics and SPSS software.
I am sincerely grateful to all of my friends and strangers whose efforts helped spread the survey reaching a sample size that otherwise I would never have reached.
Lastly I want to express my grateful thanks to all the people that responded to the online questionnaire, for the time spent and honest answers. Without them, I would not have any research material to study and to provide some relevant conclusions.
Page: 5 of 64
1. Introduction
One of the most prominent concerns in the entrepreneurial setting is the uncertainty involved in business decisions (Hebert and Link, 1989). Uncertainty is described as a gap of information regarding the understanding or knowledge of an event (ISO Standard 31000:2009). Generally, mangers possess less information then they would like, leading to limited foresight about the future (Surowiecki, 2004). Firms face uncertainty when creating a new product or just improving an existing one (Crawford, Aguinis, Lichtenstein, Davidsson and McKelvey, 2015). The success of a business depends therefore of an efficient decision-making mechanism, in which its efficiency is understood by the way it performs under the conditions of uncertainty (Surowiecki, 2004).
An efficient decision-making mechanism has been recognized in the potentially large and unknown population, i.e., the crowd (Afuah and Tucci, 2012; Howe, 2006; Jeppesen and Frederiksen, 2006; Terwiesch and Ulrich, 2009). The crowd is defined as a “remarkably intelligent and often smarter than the smartest people in them” (Surowiecki, 2004). Technological advances make access to the “wisdom of the crowd”1, easier than ever, thanks to the growing importance of the internet. Firms can, due to this empowerment of the internet, built online communities, encourage customer-involvement and gain valuable customer feedback that can ultimately improve the company ability to innovate2 and anticipate future consumer needs (Prahalad and Ramaswamy, 2002). Companies such as Google, Slashdot and Wikipedia (Surowiecki, 2004); Dell, Lego, Starbucks, or Threadless (Poetz and Schreier, 2012; Stephen, Zubcsek and Goldenberg, 2016), show how successful crowd strategies can be.
The term "crowdsourcing" was used for the first time in 2005 by Jeff Howe and Mark Robison, defined as “representing the act of a company or institution taking a function once performed by employees and outsourcing it to an undefined (and generally large) network of people in the form of an open call. This can take the form of peer-production (when the job is
1 “Wisdom of the crowd” it is a concept brought by Surowiecki. (2004), in its book: The Wisdom of Crowds:
Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business.
2 Innovation may start from using new knowledge or reusing and combining existing knowledge (Anderson,
Page: 6 of 64 performed collaboratively), but is also often undertaken by sole individuals. The crucial prerequisite is the use of the open call format and the large network of potential laborers." After this the company, can select the best of the resulting ideas and transform them into new products or just improve the company existing products (Nishikawa, Schreier, Fuchs and Ogawa, 2017). So consequently, crowdsourcing consists in a fundamental key for the innovation process of a company. As companies, are increasingly search for new product ideas outside their own boundaries (Franke, Poetz and Schreier, 2014; Von Hippel, 2005). We live in a world of distributed knowledge; companies should not only rely on their own research and development (Chesbrough, 2003). The innovation concept is therefore the “implementation and execution of creative ideas” (Amabile, 1996; Klein and Sorra, 1996; Shalley, Hitt and Zhou, 2015), making these creative ideas in to new products and offering new services, or adding new value to existing ones (Cassey and Guing, 2007). Consequently, crowdsourcing operates by helping the firms to identify those creative ideas, which will lead them in end to innovate. Crowdsourcing like this is also very useful because it helps the companies to adapt, if necessary, before potentially wasting money (Ebel, Bretschneider and Leimeister, 2016).
Innovation scholars already begun to study the effect of labelling products as a result of crowdsourcing (user-designed3) at the point of purchase (POP), as a way to communicate a product. Scholars such as Nishikawa, Schreier and Ogawa (2012), showed that crowdsourcing is an “innovation tool” (to identify promising ideas for new products), and there is a point of differentiation in labelling products, as user-designed. Their research showed that labelling products as user-design had important effects in the bottom line. Muji (a Japanese consumer goods firm, which they studied) increased up to 20% the product´s market performance. The authors also showed that customers prefer user-designed, new products at the POP (vs. created by company-internal designers)4. In short, crowdsourcing (or, more generally,
3 To simplify: a product that has resulted from crowdsourcing, which is the definition of user-designed provide by von Hippel. (2005) “a user-designed product refers to one that has been created by a user, who resides outside the contractual boundaries of the firm; a user refers to an individual such as a consumer or community member who primarily realizes product benefits by using it”.
4 To simplify: a product or ideas that were generated by the firm’s professionals, which is the definition of created by company-internal designers provided by Fuchs, C., Prandelli, E., Schreier, M., & Dahl, D. W. (2013): “define a product created by a company-internal product designer as one where the original design is conceived by a professional employed by the underlying brand”.
Page: 7 of 64 customer-centric innovation) might not only constitute a promising route to better new products but also help marketers set their products apart from the competition (Nishikawa et al., 2017; Albors, Ramos and Hervas, 2008; Schenk and Guittard, 2011).
Meanwhile research also pointed to the negative effect of labelling products as a result of user-designed in areas such as luxury fashion industry5 and pharmaceutical industry. In the case of the luxury fashion industry a decrease in demand is showed when products are user labeled (Fuchs, Prandelli, Schreier and Dahl, 2013; Moreau and Herd, 2010). In the case pharmaceutical industry, crowdsourcing cannot be applied, once new medicines result from scientific research processes and not from public suggestions (Nishikawa et al., 2017). Moreover, while fashion brands with their company-internal experts have continuously demonstrated their skills and ability to conceive high quality designs, users might be perceived by consumers to lack the related expertise (Fuchs et al., 2013). In this context user design also backfires because user-designed items provide the wrong signal in the marketplace. Indeed, in certain situations (e.g. luxury items) user-design fails to provide consumers’ agentic feelings (e.g., “I am better than others”) (Fuchs et al., 2013).
The aim of this thesis is to understand the value of crowdsourcing as a labelling strategy when compared with other related labelling strategies that can be used at the POP, such as consumer reviews and top sales, as suggested by Nishikawa et al. (2017). Still considering the findings of Fuchs et al. (2013) study in labeling products as user designed backfires, it’s taking in consideration his suggestion of labeling products as celebrity endorsers, as an influence in the perception of quality by providing agentic feelings to the consumers. This was already demonstrated in the context of the luxury fashion industry, but the interest now is to understand whether such effect also occurs outside the luxury setting. Such understanding maybe of relevance to marketing managers outside the fashion industry (Fuchs et al. 2013). This research aims to understand if there is a more effective way for a company to promote/communicate new products, when comparing the following labeling strategies:
5 Luxury is derived from the Latin word “luxus” which translates into “excess;” luxury products in general thus
refer to products leading to a condition of abundance, something that adds to pleasure or comfort but is not absolutely necessary (Encyclopedia Britannica).
Page: 8 of 64 crowdsourcing, consumer reviews, top sales or public figure endorsement. Consumer reviews are important as the Bright Local’s 2017 Local Consumer Review revealed that 88 percent of consumers trust online reviews as much as a personal recommendation). Top sale product is likely to be a reflection of ones needs to comply with social norms and group belonging: consumers purchase action depends on what the other consumers are choosing (McFerran, Dahl, Fitzsimons and Morales, 2009). Finally public figure endorsement provides the agentic feelings and an endorsement contract with a celebrity is proven to increases sales (Crutchfield, 2010).
To test our idea, we conducted an experimental study. An online-survey was made to simulate a POP strategy label to understand which one generates more trust in the consumer and why. Taking into account that quality6 and innovation perceptions are important factors, in this type of decisions. Quality is important in the sense that consumer perceived quality does influence purchase intention and perceived value of a product (González-Benito and Martos-Partal, 2012; Calvo-Porrala and Lévy-Mangin, 2017). Innovation in the sense that realizing customers’ needs and wants with new characteristics’ through creating new product pattern develops purchase intentions, to purchase the product among consumers (Kleinschmidt and Cooper, 1991). Moreover, product features are the major decision variable used by the marketer to influence the product evaluations and the purchase behaviors of potential customers (Seng and Ping, 2016). To effectively make decisions regarding these variables, marketers seek knowledge about how consumers use product attribute information in the evaluations of products (Chang and Wildt, 1994). When consumers are asked why they have recently purchased products, they mention price and performance (product features) as reasons, which are measures of overall value (Hoyer, 1984). As such, we seek also to understand which of the label strategies is more associated with these quality and innovation concepts perceptions. Our conceptual model is summarized in Figure 1.
The chapters of this thesis are organized as follows: in chapter two the existing literature on crowdsourcing, customer reviews, top sales and public figure endorsement concepts are
6 “Quality is perceived differently by different people. Yet, everyone understands what is meant by “quality.” In
a manufactured product, the customer as a user recognizes the quality of fit, finish, appearance, function, and performance. The quality of service may be rated based on the degree of satisfaction by the customer receiving the service. Quality is the degree to which performance meets expectations” (Chandrupatla. 2009).
Page: 9 of 64 presented and important features are then identified. While based on the literature, the key relationships are hypothesized. Then in chapter three the experiment, its objective and its method is presented. In chapter four the results of the experiment are highlighted, followed by a discussion and academic and managerial implications in chapter five. Lastly, the conclusions and limitations are stated and areas of possible future research are suggested.
Figure 1 the conceptual model.
2. Literature Review
2.1 Definitions
2.1.1 Crowdsourcing
New products failure is a big concern in any firm. According to Ogawa and Piller (2006) this failure occurs not due to technical issues but because is extremely challenging to produce something that customers will want. To fight this several researchers pointed out crowdsourcing as one efficient solution.
The word crowdsourcing is a compound contraction of crowd and outsourcing. Thus, crowdsourcing means outsourcing to the crowd. Crowdsourcing is a form of outsourcing not directed to other companies but to the crowd (Schenk et al., 2011).The crowd can be defined as a large set of anonymous individuals, due to this anonymity; individuals cannot be individually identified or recognized. Implicit here is the idea that a firm cannot “build its own crowd”. Moreover, the crowd is generally composed of heterogeneous individuals. In particular, a crowd may be composed of scientists and experts in various fields, but also of
New Product Communications: - Crowdsourcing
- Consumer Reviews - Top Sales
- Public Figure Endorsement
Purchase Intention: - Product Purchase
Intention Scale - Attitude Towards the
Firm Scale Perception of:
- Innovation - Quality
Page: 10 of 64 novices (Nambissan and Sawhney, 2007). So, crowdsourcing generally involves three categories of actors: the individual’s (who form the crowd); the companies directly benefitting from the crowd input; and the intermediation platform, who is building a link between the crowd and the companies (Schenk et al., 2011).
Crowdsourcing, therefore, is a new trend towards integrating consumers/users (the crowd) in the productive processes, by responding to crowdsourcing/open calls (to give feedback), mostly via an Internet platform. This, allows firms not only to reach a greater number of individuals but also to create new relationships by persuading new customers to work for the firm or for its users (Kleemann, Rieder and Voss, 2008). After this call the firm selects the best ideas and converts them into new products or just in simple improvements of existing products already offered by the firm (Nishikawa et al., 2017).
The efficiency in crowdsourcing relies in the fact that is an easy, fast and affordable way to mobilize large numbers of people (Garrigos-Simon, Gil-Pechuán, Estelles-Miguel, 2015; Simperl, 2015) in order to get access to expertise which a firm never knew it needed and never had before a way to find it (McNeal, 2013). Moreover, what gives strength to this efficiency is that this crowdsourcing wisdom arises not from an average of solutions but from the aggregation of all them (Brabham, 2008; Surowiecki, 2014). It is from this combination (of all solutions/ideas) that the best one (solution) is born, and it’s what makes crowdsourcing thrive.
The main advantages of crowdsourcing are the following: first, by employing users instead of professional designers, in the innovation process, these user-designed ideas have the same features or characteristics as those desired by the segment to which companies are trying to sell to. In this way, they can more easily read the potential customers’ needs and wants, leading to higher changes of having successful innovation’s (Nishikawa et al., 2017). The second advantage is that by using these user-designed ideas as labelling, customers display higher intentions to purchase products at the POP, because they are made aware that products were created by similar others (the users) (Dahl, Fuchs, and Schreier, 2014), resulting in an increase in brand loyalty (Nishikawa et al., 2017; Sawhney, Verona, and Prandelli, 2005) and consequently in a higher demand for the firm’s products (Fuchs et al., 2013). Moreover, it helps marketers differentiate themselves from competitors, by offering them a competitive advantage (Nishikawa et al., 2017). The third advantage consists in the fact that companies
Page: 11 of 64 which use user-designed ideas are associated with higher innovation abilities, when compared to firms that internally create their products (Lude, Hauck, Prügl, Linzmajer, 2016). The fourth advantage states that diversity is believed to generate more desirable products (Surowiecki, 2004). This happens because user-designed ideas involve a greater number of people in its process of creation, than number of people which constitutes the professional designers of a company. The fifth and last advantage is that by using user-designed ideas the creations that happen from it were less constrained by a company rules and goals. This is translated in more freedom to be creative and innovate, and therefore in return the products created were more desirable to the consumers (Schreier, Fuchs, and Dahl, 2012).
Inspired by Nishikawa et al., (2017) and Fuchs et al. (2013) investigations in order however to really understand the true value of crowdsourcing labeling as a quality and innovation signal, to promote consumer demand for a product, is important to compare it with other labeling strategies that can also be used at the POP, as mentioned before. These other labeling strategies where identified as consumer reviews, top sales and public figure endorsement.
2.1.2 Consumer Reviews
Every company if it has customers, it provides consciously or not a customer experience. The experience can be good, bad or indifferent (Kotler and Armstrong, 1996). No company can completely control the experiences which they give to their customers, because experiences involve perception, emotion and unexpected behaviors on the parts of customers (Richardson, 2010). Despite this lack of control companies can try to plan ways to mitigate this when considering the experiences, they want to create (Richardson, 2010).
Customers are conscious that their feedback matters (Rassega, Troisi, Torre, Cucino, Santoro and Prudente, 2015). Turning their testimonials into a tool rather than a threat is important. To do this, companies must: understand what customers are saying; understand where they are saying it; and (most important) understanding why they are saying it. But, today there is a sea of online feedback and this complicates the identification of the valuable reviews (Hicks, 2012).
Page: 12 of 64 Customers are responsive to other customer reviews. Customer reviews works as a social proof, that a business is legitimate and credible and helps to prevent other potential customers to fall for a product or service that will not maximize their needs and wants (Bookbinder, 2017). BrightLocal survey showed that 88 percent of consumers trust online reviews as much as a personal recommendation. This finding suggests that if there is a product or service that is not offering user reviews (or ignoring them as a potential marketing opportunity) is akin to alienating 88 percent of your buying population (DeMers, 2015). Also, there is an increasing expectation gap as business struggle to keep pace with more informed, more connected and more demanding consumer (Perkins and Fenech, 2014), as consequence consumers, now more than ever, are demanding. Thus, if a product has positive or bad review it holds a significant effect on the behavior of the rest of the audience (Sharma and Rehman, 2012; DeMers, 2015). At the same time, customer reviews increase and stimulates conversation; this can result in the elimination of potential uncertainty customers have when purchasing a new product (Charlton, 2015).
When they are looking for product information consumers prefer “independent” sources such Amazon or TripAdvisor, in opposition to other nonneutral sources of information, such advertisements, brochures, company Web sites, and salespeople (Kotler et al., 1996). TripAdvisor is therefore a good example of the influence and power of consumer reviews. TripAdvisor is the biggest travel reviews website in the world (Vanderbilt, 2015). Daily Mail conducted an independent study of TripAdvisor and concluded that user-generated content was “directly related to £2 billion of tourism spending in the UK” (Kitching, 2016). Reviews written by previous visitors to places in the UK have influenced 8.7 million trips by tourist (Kitching, 2016). Also, according to Vanderbilt (2015) study: “hotel owners who reply to comments are 20% more likely to get bookings and hotels can raise prices by 11% to reflect their reputation on the site”. Barrie (2015) concluded as well that “every positive percentage point a place rises up the tables in TripAdvisor, revenue per room at locations increases by 1.4%”.
A survey by Forrested of over 2100 travelers commissioned by TripAdvisor found that 81% of traveler’s suggest reviews are important, while 3% suggest they were not, however, almost half of the respondents said they wouldn’t book or look at a hotel unless it had reviews. Thus, the influence and feedback from costumers is essential for hotels survival and reputation (Vidigal, 2017). TripAdvisor influences and generates increased travel spend. In reality,
Page: 13 of 64 hotels with good rankings and reviews can enjoy higher, booking, higher average daily rate and revenue (Lyle, 2015).
To conclude, customer reviews influence the business’s digital presence and in turn, sales (Kim and Srivastava, 2007). Reviews not only show consumers that a brand is trustworthy but they also have the potential to improve the business’s SEO (Bookbinder, 2017). Overall, customer reviews are a vital piece in the process of convincing prospects that a product or service is better than the competition and that truly offers the value that firms are selling. For buyers, it is a vote of confidence that the purchase decision, which is being made, will be a good one, and if this confidence it’s proved to be right, it will make them, later, more willing to purchase again (Bookbinder, 2017). Like this customer reviews have enormous potential to turn a passive shopper into a lifelong, loyal buyer.
2.1.3 Top Sales
A top sale consists in a product which is most popular among customers and that consequently is also the one that has being selling in larger quantities (Cambridge Dictionary). A top selling product or service itself yields, sometimes, enough profit to justify the continued existence of a firm or business (Codjia. 2000).
To understand why top sales are influential in the purchase intention of consumers is important to look at social norms and group belonging. Social influence has showed to be important in the consumption process (Bearden and Etzel, 1982; Argo, Dahl and Manchanda, 2005). Most of research has been focus on how an interactive social influence, such as salespeople, impacts the consumer (Childers and Rao, 1992). However social influence situations in purchasing are not limited only to these interactive situations, but to situations that occur without interaction, called noninteractive social situations: “includes events where a social entity is physically present during consumption but is not involved nor attempts to engage the consumer in any way (e.g., other shoppers in a grocery aisle or a fellow audience member at the theater)” (Argo et al., 2005). In this context, a top sales presents a noninteractive social situation.
Page: 14 of 64 The consumer attitudes depend on socialization factors (e.g., peers) (Taylor, Lewin and Strutton, 2011). According to Wang, Yu and Wei (2012) online consumer socialization through peer communication effects purchasing decision in two ways: directly (conformity with peers) and indirectly by reinforcing product involvement. People are impacted by the “real, implied, or imagined presence or action of a social presence (e.g., another person or group of people) (Argo et al., 2005). A top sales, refers as mentioned before, to the product or a service, which consumers are buying the most. The social other is purposeful in the social behavior that influences the actions of the consumer (Dahl, 2013). Consumers purchase action depends on what the other consumers are choosing (McFerran et al., 2009). Knowing that most people are buying, for example a specific model of a t-shirt, in order to be part of the group, leads consumers to purchase the shirt, even if the consumer may not even need a t-shirt (Kotler et al., 1996).
Another important factor to understand why top sales influence consumers purchase intention is that consumers perceived top sales as a profitable sales volume, meaning these products are associated with long-term customer satisfaction (Schiffman and Kanuk 2004). So when a product is presented as top sales, the message is that this product has quality, because it satisfies the needs and wants of the consumer (Akdeniz, Calantone and Voorhees, 2014). Moreover, the product is perceived as possessing intrinsic quality—related with internal product characteristics - being the perceived product quality a key determinant in building and maintaining customer loyalty (Brakus, Schmitt and Zarantonello, 2009; Pan, Sheng and Xie, 2012). A study conducted by Garrido-Morgado, González-Benito and Martos-Partal (2016) showed that the dimensions of quality perceptions moderated the influence of displays and advertising flyers on sales, and that intrinsic quality perception improved the effect of advertising flyers, which in turn are more closely related to systematic decision processing.
2.1.4 Public Figure Endorsement
Recent research showed that user-design labeling strategy does not benefit all products. In fact, there are negative effects in such labeling (Fuchs et al., 2013). The luxury fashion industry and the pharmaceutical industry seemed to be the most affected ones, with these negative effects (Nishikawa et al., 2017; Fuchs et al., 2013; Moreau and Herd. 2010). Nonetheless fashion brands recognize the promises of user-design. For example, the handbag
Page: 15 of 64 brand Coach, invited users to participate in a “Design a Coach Tote” initiative which resulted in 3,000 user-designs, the best of which were manufactured by the brand (Fuchs et al., 2013). The real motivation however for the luxury fashion industry to use crowdsourcing is not only to get better products, but to target the broader mass of consumers, by seducing them to participate in some online voting process, with the aim of increase the consumer’s involvement and ultimately their commitment to buy from the underlying brand (Fuchs, Prandelli and Schreier, 2010; Schau, Muniz and Arnould, 2009).
Meanwhile as it is known the fashion industry has always distanced itself from the mass consumers (Kapferer and Bastien, 2009). This is explained by the fact that being close to the mass consumers does not help but harms luxury fashion brands, due to user-design stands far from the high status signaling (Fuchs et al., 2013). Moreover, user-designed ideas fail in create the feeling of high status/ agentic feelings: “being advantaged, superior and worthy compared to others” (Locke, 2003).
The two useful lessons to be considered with the case of the luxury fashion industry are the following: first, there are still consumers who prefer to buy a product developed by the internal designers of a brand versus user-designed ideas. This occurs for consumers who perceived higher quality only when products are created by the company-internal professionals, i.e., product designers (Dubois, Laurent and Czellar, 2001). Often users are perceived to lack the expertise to create premium products (because it will not have enough quality) (Fuchs et al., 2013). The second lesson lies in the fact that some consumers, want to signal themselves apart from others (the mass consumers) (Fournier, 1998; Rucker and Galinsky, 2008). To achieve this, these consumers, do not want to be in conformity with the crowd, but above it. These consumers want to create social distance by a boost of agentic feelings (Fuchs et al., 2013).
In this context the strategy of using a public figure endorsement is introduced, once while still being able to provide agentic feelings, can in some circumstances be presented as not having resorted to company internal designers, in the cases when the endorser actively participates in the product’s design. This labeling strategy may range from the celebrity only lending their name or image to a product or campaign, not being really involved in the design of the product (Passikoff, 2013), to case which involves the carefully participation of one or some celebrities who are specially invited to participate in the product design (Fuchs et al., 2013).
Page: 16 of 64 The involvement of famous celebrities’ users is shown to be useful, because celebrities are perceived by others as possessing status and thus owing some social distance from mass consumers, i.e. they possess agentic feelings (Okonkwo, 2007). Therefore, these celebrities’ endorsers activate perceptions of design quality in the product that enable those agentic feelings to the consumers that purchase that product (Okonkwo, 2007).
To conclude public figure endorsement has a simple logic: people idolize celebrities, so when famous people are seen in advertisements promoting a new product, audiences are prompted to buy that product, either subliminally or directly (Olenski, 2016). Moreover, one study by Chaudhary and Asthana (2015) found that celebrity endorsers do not necessarily influence consumer brand loyalty; endorsements are instead a powerful and useful tool that magnifies the effect of a campaign. Crutchfield (2010) showed that, a brand that does an endorsement contract with a celebrity or an athlete can see their stock rise up as soon as the news is made public.
3. Hypotheses
Arguments were found in favor of each strategy and as such it is not possible to exactly point the direction of the hypotheses, meaning that apriority we cannot favor one of the following hypotheses.
3.1 Crowdsourcing Hypothesis
Crowdsourcing when compared with other labeling strategies may increase the purchase intention of the consumer. Several reasons can be considered. First, in the digital landscape and in the globalized competitive arena, companies alone can no longer respond to this complex growing environment of innovation (Chesbrough, 2003; Oldham and Da Silva, 2015), to find successful ways to satisfy the needs and want of its users. Consequently, products labeled as company design or as public figure endorsement may not hold enough innovation to attract consumers. In the case of public endorsement this may happen because sometimes the celebrities’ endorsers only lend their name to the product or just appear in a
Page: 17 of 64 couple of ads (Passikoff, 2013). As such celebrities are not involved the process of generating ideas for the product, meaning that the product is still created as company design creation. Secondly, in crowdsourcing a company can select then the best of the ideas given by the users (Schenk et al., 2011; Nishikawa et al., 2017), meaning that the company still holds control about the process of selection, of what are the useful and convenient ideas, in consumer reviews there is no control whatsoever about the information produced, depends on each personal experience and perception of each user of the product (Richardson, 2010; Paljug, 2017).
Finally, while crowdsourcing has in its foundation the creation or the improvement of a product (Cassey et al., 2007; Nishikawa et al., 2017) a top sales it´s only rephrasing an existing product with no alterations. As such, the following relationship is hypothesized:
H1: Labeling a product, as crowdsourcing increases the purchase intention of the consumer, when compared with the strategies of consumer reviews, top-sales and public figure endorsement.
3.2 Consumer Review Hypothesis
Consumer reviews when compared with other labeling strategies may increase the purchase intention of the consumer. Since, they enable a deeper knowledge and feedback about a product uses and features. Consumer reviews arises from experience, after the service or the product has been created (Hicks, 2012), thus focusing on the outcome. While crowdsourcing provides feedback about the creation of a new product or the improvement of an existing one (Cassey et al., 2007) thus focusing on the inputs. Thus, crowdsource alone doesn´t provide feedback or results after the product is launched, to show to consumers if product is worthy or not for maximizing their needs and wants (Bookbinder, 2017).
Second, a product presented by customer reviews includes in its creation positive and negative feedback, while a top sales product includes only the most positive side of a product (Kotler et al., 1996), a customer review labeling strategy can be seen as a more complete, than a top sales one.
Page: 18 of 64 Third, customer reviews encompass a wider range of consumer opinions (Rassega, et al., 2015), which can include celebrities opinion as well (as they can even leave their comment in an anonymous way if they don´t want to expose themselves), while public figure endorsement includes only the recommendation of celebrities. Therefore, the following relationship is hypothesized:
H2: Labeling a product, as consumer reviews increases the purchase intention of the consumer, when compared with the strategies of crowdsourcing, top-sales and public figure endorsement.
3.3 Top Sales Hypothesis
Top sales when compared with other labeling strategies may increase the purchase intention of the consumer due to: first, a top sales shows exactly what are consumers buying (Argo et al., 2014), successful companies as Mango, H&M and Zara expose in their websites a category called “best sellers”, where it’s showed their most selling products of each collection. While crowdsourcing is used to create, or improve a product but doesn´t show to the consumers (who want to belong to the group) (Kotler et al., 1996), if the product is being after launched consumed by their peers. Also, while companies exhibit their top sale products, in their stores, websites and publicity campaigns, not all companies display their products as crowdsourced ones, as it can backfires (the case of the luxury products) (Nishikawa et al., 2017; Fuchs et al., 2013; Moreau et al., 2010).
Second, top sales are associated with positive information about a product (Cambridge Dictionary), while customer review due to contain negative and positive information about a product (Kotler et al., 1996), the negative information even if is smaller than the positive one, can still shake the purchase intention of the consumer for the product, sometimes unfairly.
Third, consumers purchase intention depends on what the other consumers are choosing (McFerran et al., 2009). As a top sales includes what the majority of consumers are choosing, public figure endorsement includes only what a celebrity or celebrities are choosing (Olenski, 2016). As such, the following relationship is hypothesized:
Page: 19 of 64
H3: Labeling a product, as top sales increase the purchase intention of the consumer, when compared with the strategies of crowdsourcing, customer reviews and public figure endorsement.
3.4 Public Figure Endorsement Hypothesis
Public figure endorsement when compared with other labeling strategies may increase the purchase intention of the consumer due to: not all consumers want a product created by crowdsourcing, because they perceive it as not holding enough expertise, to be able to create a product with quality (Dubois et al., 2001; Fuchs et al., 2013) and also some consumers wish to distance themselves from the mass crowd to reach agentic feeling (Fournier, 1998; Rucker et al., 2009; Fuchs et al., 2013).
Second, due to customer review englobes reviews from anyone and public figure endorsement only includes the careful selection of certain users (the celebrities) (Okonkwo, 2007), consumers may have perceived the feedback from these celebrities as more skilled/selected, due to these celebrities holding status which is worldwide recognized (this is why they are celebrities and not common people).
Third, is proved that an endorsement contract with a celebrity increases sales (Crutchfield, 2010) and also as people idolize celebrities and wish to be as them (Olenski, 2016), so ultimately a celebrity endorsement can lead a product to became a top sale. As such, the following relationship is hypothesized:
H4: Labeling a product, as public figure endorsement increases the purchase intention of the consumer, when compared with the strategies of crowdsourcing, customer review and top sales.
Page: 20 of 64
4. Methodology
4.1 Objectives and Overview
To test the above presented hypotheses, an online survey was conducted. The aim was to find evidence of which of the four labeling strategies (crowdsourcing, customer reviews, top sales and public figure endorsement) could lead to a higher purchasing intention, when these strategies are compared among each other.
There were 288 respondents that took part in the study (61,8% female, mean age= 23). This study followed a four-group design experiment (design mode: crowdsourcing, top sales, public figure endorsement and customer reviews).
4.2 Method
An online survey was conducted, distributed on Qualtrics. This method was chosen because it was important to have control over the research environment, as some researchers argue this is the best path of creating an experimental research (Charness, Gneezy and Kuhn, 2012). The advantages consist in the fact that this is a simple experiment and the random assignment between groups is easier to be achieved (Charness et al., 2012). Moreover, an online survey provides a non-intimidating environment; therefore, respondents can be more likely to provide open and honest feedback (DeFranzo, 2012). The data gathered was then analyzed using SPSS (Social Package for the Social Sciences) software.
4.3 Survey Design
The survey was originally designed in Portuguese (Appendix I) but an English version was also created (Appendix II). To start, the respondents were presented with the survey’s goal (a study to evaluate their perception about a new product to be released in the market) next the respondents were informed about which was the new product (a cake). The choice of this product, a cake, was made, because a unisex product was sought, as well as a product whose purchase intention would not be sensible to the age factor, to facilitate the process of data collection. If, for instance, a piece of clothing had been chosen, like a t-shirt, it would require two surveys: one for men and another for women. Even so, bias could also emerge due to the
Page: 21 of 64 age factor: if the model of t-shirt to be chosen would please an elder population it probably wouldn´t please a younger one and reversely.
In the first question, respondents had to answer about their taste for cakes. Then participants were randomly presented, with the communication strategies (crowdsourcing, customer reviews, top sales and public figure endorsement). All participants read the same introduction: “Imagine that you want to buy a cake for a dinner that you will host at your house.” Then, depending on the labeling strategy being surveyed, the way the cake was presented to the consumer was different. In the crowdsourcing scenario, the new cake was presented to the participant as having been created by a national pastry, which used their online platform to collect ideas and feedback about their user’s favorite recipes, to create this new product. In the customer review version, the new cake presented to the participant had been created from consumer’s comments and criticisms gathered from several pastry blogs and websites, about the most recent cakes they purchased, were their most positive reviews pointed out to this new cake released in the market. In the top sales version, the new cake presented to the participant had been created by a national pastry which used as inspiration for this cake their all times best-selling recipes. Finally, in the public figure endorsement version, the new cake presented to the participant had been created by a national pastry that, for the development of this cake, hired a famous Pastry Chef.
Then, respondents were asked how strongly they believe users were involved in the product conception process. Only after this question a picture of the cake was displayed, showing a multiple-sliced cake which seeks to join several tastes in only one cake, to ensure greater satisfaction in one single product. The same cake picture was shown to all respondents independently of which version of the second question they were presented. The following questions of the survey tried to capture the perceived innovation ability, the willingness to try the product, the willingness to visit a pastry with this product, the intention of purchase the product and the intention to recommend this product to others. Before leaving, respondents were asked to fill out statistic information about themselves, which included the gender, age, nationality, professional situation, level of education and their average monthly income.
Page: 22 of 64
4.4 The Measures
The questions in the survey allowed to capture the perceived innovation, perceived quality and users involvement, to influence the consumer’s purchase intention towards a product, presented with one of the specific labeling strategies that are being studied (crowdsourcing, customer reviews, top sales and public figure endorsement). For analysis purposes a 7-points Likert scale from 1 to 7 was chosen, to evaluate the intensity of relationship between the variables that were been measured, in the questions made. As this scale enables an easier capture of the feelings and opinions of the respondents, where each number/level poses only one characteristic, makes clear what the respondent is responding to (Bowling, 1997; Burns and Grove, 1997). A Likert-type scale assumes also that the strength/intensity of the experience is linear,
i
.e. on a continuum from strongly agree to strongly disagree, and makes the assumption that attitudes can be measured (McLeod, 2008).Construct Items
Please state on a scale of 1 to 7, to which degree you consider there is innovation in this cake: 1. Not Innovative [1:7],
2. Very Little Innovative [1:7], 3. Little Innovation [1:7], 4. Indifferent [1:7],
5. Reasonably Innovative [1:7], 6. Innovative [1:7],
7. Completly Innovative [1:7].
Please state on a scale of 1 to 7, which do you think is the cake´s degree of quality:
1. None [1:7], 2. Vey Poor [1:7], 3. Poor [1:7], 4. Indifferent [1:7], 5. Reasonable [1:7], 6. High [1:7], 7. The Highest [1:7].
Please state on a scale of 1 to 7, which, do you think, was the degree of users involvement in the creation of this new cake:
1. Nonexistent [1:7], 2. Vey Weak [1:7], 3. Weak [1:7], 4. Indifferent [1:7], 5. Reasonable [1:7], 6. Strong [1:7], 7. Total [1:7]. Table 1 Measures Users Involvement Perceived Innovation Perceived Quality Measures
Page: 23 of 64 To better analyze and interpret the results, the purchase intention was separated in two different scales, by the aggregation of some variables. This allowed to measure product and firm attitudes (Dahl et al., 2014). When deciding about a purchase, consumers are influenced by the brand/firm (denominated attitude towards the firm) (Forte and Lamont, 1998) and the product (denominated product purchase intention) (Duboiset al., 2001; Fuchs et al., 2013). To use the data gathered about the product purchase intention (scale), we tested for statistical reliability. As such, a reliability test was performed. Below, in Table 2, it can be observed, that the constructed scale for product purchase intention has Cronbach’s alpha coefficient of α= 0.927. This value of alpha proved internal consistency, meaning that this scale can be considered reliable, therefore is justifiable to interpret scores that have been aggregated together. Also, high item total correlations are presented in Table 2. Moreover, this leads to conclude that if any of these four variables (which constitute the scale: product purchase intention) was removed, the result would be translated into a lower Cronbach´s alpha.
The scale attitude towards the firm, is composed only by one variable (intention to recommend a firm with this product), as is shown in Table 2. Therefore, reliability cannot be assessed for single item measures. However, it has been recently demonstrated, by several authors, that single-item measures can be a viable alternative to multi-item scales (Drolet and Morrison 2001; Bergkvist and Rossiter 2007; Shamir and Kark, 2004). Concerning internal consistency reliability, several authors showed acceptable reliability values for single-item scales (e.g., see Ginns and Barrie, 2004; Kwon and Trail, 2005).
A single-item measure is considered to be an individual measure or indicator (Bagozzi and Heatherton, 1994). Simplicity, brevity and global measurement are the advantages of using single-item scales (Kwon et al., 2005). Fuchs and Diamantopoulos (2009) paper offered criteria for assessing the potential use of single-item measures. About the criteria, first, in the
Scale Scale Items Corrected Item Total Correlations
Willingness to Try the Product 0.80 Intention to Recommend the Product 0.82 Willingness to Buy the Product 0.85 Willingness to move to Buy the Product 0.86
Atittude Towards the Firm: Intention to Recommend a Firm with this Product
---Table 2 Measures for Purchase Intention
Product Purchase Intention (Ca = .927):
Page: 24 of 64 nature of the construct, it is relevant whether the focal construct is concrete7 or abstract (Rossiter, 2002). One example given by the researchers of this concrete constructs is the buying intention (Fuchs et al., 2009), which is ultimately what we are trying to analyze with the creation of this scale. Moreover, when a construct is concrete, the use of single item measures is considered reasonable; due to the measurement error is more prevalent for abstract versus concrete concepts (Rossiter el al., 2002 and Fuchs et al., 2009). Secondly, in the research objectives, the single-item global rating method may be useful if the goal of a study is to gain an understanding for the general nature of construct (Lee et al., 2000). Thirdly, concerning sampling considerations, if a measure is to be administered to a wide range of different populations, the use of single-item measures has certain advantages (Fuchs et al., 2009). An advantage is that it can be given to numerous people (Gorsuch and McPherson, 1989). Single-item measures are flexible. So, taking into account the difficulty of obtaining large sample sizes in surveys, due to the lack of willingness of sacrifice time to do them, leads to the necessity of reducing the length of construct measures (Fuchs et al., 2009). “As a rule of thumb, there should be at least ten times as many respondents as items or, in cases where a large number of items are used, at least five respondents per item” (Nunnally, 1967; Peter, 1979). This rule of thumb was respected in the conducted survey once there are more than five respondents per item. In fact, for the single-item which measures and constitutes the scale “attitude towards the firm”, a total of 288 responses were obtained (Appendix III).
7Being concrete referred to objects and their characteristics which are perceived as similar by all raters and also they understand that there is only one characteristic being referred to when the attribute is posed, as in a questionnaire, in the context of the to-be rated object) (Rossiter, 2002).
Page: 25 of 64
5. Results
5.1 Description of the Sample
The data of this study was collected, as mentioned above, through a single survey. There was a total of 288 responses to the survey (N = 288) (Appendix IV). In terms of gender the respondents were 178 women (61.8%) and 110 men (38.2%), from 15 different countries (Appendix V), where most of respondents were Portuguese. The respondent’s age ranged between 16 and 67 years old with an average age of 23.
As previously mentioned the second question of the conducted survey had four different versions (without the respondents been aware of it). Each version corresponded to a different labeling strategy. The results showed that 72 respondents completed the crowdsourcing version, 74 the customer review, 73 the top sales and 69 the public figure endorsement (Appendix VI).
5.2 Product involvement
The first question was designed to understand whether respondents were familiar with the product stimuli. The average response was M = 5.79, where 103 respondents chose scale 6 “I like it” (Appendix VII). This means that most of the respondents liked, a priori, the product used to test the hypotheses.
5
5.3 Main Analysis: testing the hypotheses
5.3.1 Results for the One Way ANOVAS
- Testing for Product Purchase Intention
In the four hypotheses (H1, H2, H3 and H4) we wanted to understand if one of the labelling strategies (crowdsourcing, customer reviews, top sales and public figure endorsement), could overlap the others in influencing the purchase intention of the consumer. One-way ANOVA was conducted on purchase intention (firm and product) to test the impact, if any, of these labelling strategies, on the purchasing intention.
Page: 26 of 64 The One way ANOVA revealed no differences in purchase intentions regarding the four labelling strategies. Regarding purchase intention towards the product (F(3,287 = 0.041); p>.05) and towards the firm (F(3,287 = 0.242); p>.05). Results meant there is no enough evidence to reject the null hypotheses and that the respondent’s means are all equal. High p-values were obtained and this translated that the data is likely with a true null.
- Testing for Perceived Innovation and Perceived Quality
In order to understand whether the different communication strategies had an effect on innovation or quality perceptions, we ran another one way ANOVA, first using as dependent variable innovation and the scenarios as the independent variables. The results indicate no differences in the perception of the level of innovation according to the label associated with the product (Mcrowdsourcing on perceived innovation = 5.51; Mcustomer reviews on perceived innovation = 5.64; Mtop sales on perceived innovation = 5.34; Mpublic figure endorsement on perceived innovation = 5.43) (F(3,287) = 0.558; p>.05).
Another analysis of variance on perceived product quality showed all scenarios were perceived with the same quality (Mcrowdsourcing on perceived quality = 5.35; Mcustomer reviews on perceived quality = 5.47; Mtop sales on perceived quality = 5.37; Mpublic figure endorsement on perceived quality = 5.41) (F(3,287) = 0.134; p>.05). Therefore consumers did not perceived different quality to the product depending on the labeling strategy used.
Scale Source F df Sig
Product Purchase Intention Scenarios ,041 3 ,989
Attitude Towards the Firm Scenarios ,242 3 ,867
Total 287
Table 3 Scenarios on the Purchase Intention
Dependent Variable = Purchase Intention Two Scales TABLE 3
Page: 27 of 64 Therefore we reject all four hypotheses.
Scenarios Mean Std. D. N
Crowdsourcing 5,51 1,538 72
Customer Reviews 5,64 1,245 74
Top Sales 5,34 1,465 73
Public Figure Endorsement 5,43 1,43 69
Total 5,48 1,419 288
Table 4 Mean Scenarios on Perceived Innovation
Dependent Variable = Perceived Innovation TABLE 4
Source F df Sig
Scenarios ,558 3 ,643
Total 287
Table 4.1 Scenarios on the Perceived Innovation
Dependent Variable = Perceived Innovation TABLE 4.1
Scenarios Mean Std. D. N
Crowdsourcing 5,35 1,313 72
Customer Reviews 5,47 1,185 74
Top Sales 5,37 1,275 73
Public Figure Endorsement 5,41 1,365 69
Total ´5,40 1,278 288
Table 5 Mean Scenarios on Perceived Quality
TABLE 5
Dependent Variable = Perceived Quality
Source F df Sig
Scenarios ,134 3 ,940
Total 287
Table 5.1 Scenarios on the Perceived Quality
Dependent Variable = Perceived Quality TABLE 5.1
Page: 28 of 64
5.3.2 Results for the Regression Analysis
- Testing for Purchase Intention
We confirmed out the findings with a regression analysis, to ensure that is not possible to reject the null hypotheses, due to this analysis will show if we can predict the value of a variable, based on the value of another variable.
In first regression analysis, we wanted to predict the purchase intention (as the dependent variable) using the scenarios (as the independent variable). As the purchase intention is composed by two scales: first we used the product purchase intention scale in the regression (table); and only afterwards, we used the attitude towards the firm scale (table). The variables presented a p-value higher than 0.05, which indicates that the different scenarios were not statically significant.
- The effect of labeling strategies on Perceived Innovation and Perceived Quality
Regressing our scenarios on innovation and then quality (as the dependent variables), showed, as expected, no significance confirming that labeling strategies are not influencing perceived innovation ability (
= -.042, p> .05) and perceived quality (
= .006, p> .05), which indicates that the different scenarios do not influence the level of perceived innovation or quality in the product.Source Beta t Sig
Scenarios -,020 -,337 ,737
Table 7 Coefficients: Scenarios on the Attitude Towards the Firm
Dependent Variable = Attitude Towards the Firm Coefficients
Page: 29 of 64
5.4 Further Analysis
5.4.1 Product Users Involvement
Focusing now on the variable of perceived user’s involvement, we wanted to understand if respondents perceived differences in the user’s involvement, among the different scenarios displayed, as the above main analysis couldn´t provide significant differences between the scenarios.
- Testing for differences in perceived user’s involvement among the scenarios
Resorting to one way ANOVA, in this case of measurement we used as dependent variable consumers perceiving that a certain labeling provides a strongest user involvement than others. Results show a significant difference of perception of respondents about the degree of user’s involvement. Respondents showed that public figure endorsement was perceived with the least user’s involvement (Mpublic figure endorsement perceived users involvement = 4.51), crowdsourcing was second highest (Mcrowdsourcing perceived users involvement =5.19), and top sales was perceived with the highest users involvement (Mtop sales perceived users involvement = 5.29). (F(3,287) = 4.554; p = .004). Moreover, the post hoc test, revealed that this significance existed only between crowdsourcing and public figure endorsement (p-value = .016), and top sales and public figure endorsement (p-value = .004).
Source Beta t Sig
Scenarios -,042 -,714 ,476
Table 8 Coefficients: Scenarios on the Perceived Innovation
Dependent Variable = Perceived Innovation Coefficients
TABLE 8
Source Beta t Sig
Scenarios ,006 ,104 ,918
Table 9 Coefficients: Scenarios on the Perceived Quality
Coefficients TABLE 9
Page: 30 of 64 Being top sales the only strategy, among the four strategies studied, born from the post-selling results, it therefore translates the preferences of the consumers for the products that are being most bought, once consumers perceive them as satisfying the most their needs and wants (Akdeniz et al., 2014). At the same time the consumers purchase action depends on what the other consumers are choosing (McFerran et al., 2009). Bearing this in mind, maybe is not surprising that this was the scenario which was attributed the most user involvement.
- Testing for Perceived Innovation and Perceived Quality
Next we run again an ANOVA to see if perceived innovation and perceived quality (as dependent variables) can be related with the user involvement (as independent variable). The analysis of variance (ANOVA) revealed a significant interaction effect of user’s involvement on perceived innovation (F(6,287) = 5.941; p<.001) and on perceived quality (F(6,287) = 4.118; p<.001). This makes sense, once consumers are also users and so they feel
Source F df Sig
Scenarios 4,554 3 ,004
Total 287
Table 10 Scenarios on Users Involvement
Dependent Variable = Users Involvement TABLE 10
Scenarios Sig
Crowdsourcing Customer Reviews ,854
(Mean = 5,19; Std. D. = 1,146) Top Sales ,977
Public Figure Endors. ,016*
Customer Reviews Crowdsourcing ,854
(Mean = 5,01; Std. D. = 1,419) Top Sales ,617
Public Figure Endors. ,122
Top Sales Crowdsourcing ,977
(Mean = 5,29; Std. D. = 1,230) Customer Reviews ,617
Public Figure Endors. ,004*
Public Figure Endorsement Crowdsourcing ,016*
(Mean = 4,51; Std. D. = 1,633) Customer Reviews ,122
Top Sales ,004*
Table 10.1 Scenarios on Users Involvement
*The mean difference is significance at 0.05 level
TABLE 10.1
Page: 31 of 64 connected to the user-designers and to companies which use user-designed ideas (Dahl et al., 2014).
The results of the regression analysis in this case provided evidence that the relationship between perceived user involvement and perceived innovation
= .266, p< .05 and perceived quality
= .188, p<.05 both have significance (p< .05). Reporting, perceived innovation and perceived quality, this suggests that the perceptions of higher user involvement influences how consumers rate the level of quality and innovation in the product. Higher innovation is associated once the user-designed ideas involve more people in the process of creation (Dahl et al., 2014).Source F df Sig
Users Involvement 5,941 6 ,000
Total 287
Table 11 Users Involvement on Perceived Innovation
TABLE 11
Dependent Variable = Perceived Innovation
Source F df Sig
Users Involvement 4,118 6 ,001
Total 287
Table 11.1 Users Involvement on Perceived Quality
Dependent Variable = Perceived Quality TABLE 11.1
Source Beta t Sig
Users Involvement ,266 3,921 ,000
R² (R-squared) ,051
TABLE 12 Coefficients
Dependent Variable = Perceived Innovation
Page: 32 of 64 It is then possible to conclude from the results, that the user’s involvement in the product, leads to higher perceptions of innovation and quality. The relevant question for us is whether such perception of user´s involvement is reflected in the scenarios. For that purpose below we used the scenarios of top sales and public figure endorsement, once these where the ones that showed the highest and lowest levels of user involvement to try again to explain the purchase intention and the perceptions of innovation and quality, previously addressed on the main analysis.
5.4.2 New Group: high and low user’s involvement assumption
- Testing Product Purchase Intention
Based on respondents’ reported user involvement, we defined new variables with two levels:
high user´s involvement (top sales scenario) and low user´s involvement (public figure
endorsement scenario). Then we performed a one way ANOVA, to understand if there were differences in purchase intention that could be attributed to the inferred levels of user involvement in the product development. As dependent variable we had the two scales (product purchase intention and attitude towards the firm), and as independent variable the new variables (high and low users involvement).
The One way ANOVA revealed that the variables have no statistical significance (F(1,141) = 0.102; p>.05) (F(1,141) = 0.519; p>.05). So this new created group cannot help to explain the purchase intention of the respondents.
Source Beta t Sig
Users Involvement ,188 3,244 ,001
R² (R-squared) ,035
Dependent Variable = Perceived Quality
Table 13 Coefficients: Users Involvement on Perceived Quality
TABLE 13 Coefficients
Page: 33 of 64
- Testing for Perceived Innovation and Perceived Quality
Next, we wanted to test if these new variables (high and low user involvement) could at least explain the perceived innovation and quality by the respondents. An One-way ANOVA revealed that the new variables doesn´t have a significant interaction effect on perceived innovation (F(1,141) = 0.144, p=.705) and in perceived quality (F(1,141) = .026; p=.871). Hence, we find no evidence that high and low user involvement directly moderates the relationship between perceived innovation and perceived quality.
Scale Source F df Sig
Product Purchase High and Low Users Involvement ,102 1 ,750
Intention
Attitude Towards High and Low Users Involvement ,519 1 ,473
the Firm
Total 141
Table 14 High and Low Users Involvement on the Purchase Intention
TABLE 14
Dependent Variable = Purchase Intention Two Scales
Source F df Sig
High and Low ,144 1 ,750
Users Involvement
Total 141
Table 15 High and Low Users Involvement on Perceived Innovation
Dependent Variable = Perceived Innovation TABLE 15
Source F df Sig
High and Low ,026 1 ,871
Users Involvement
Total 141
Table 15 High and Low Users Involvement on Perceived Quality
TABLE 15.1
